The role of internal cohesion in the UK Parliament and its possible effects on the Brexit negotiations

(1)

The role of internal cohesion in the UK Parliament and its possible effects on the Brexit negotiations

By

Ferdi Evers S1758713

University of Twente University of Münster

M.Sc. European Studies Double Degree Programme

Supervisors:

Dr. Shawn Donnelly, University of Twente Dr. Ivo Hernandez, University of Münster

June 2021

(2)

1

Abstract

The aim of this thesis was to take a closer look at the political cohesion in the House of Commons of the United Kingdom, measured through polarization, and how it could have affected the negotiations surrounding the withdrawal agreement. To start with a literature review was done on the effects polarization can have on negotiations. Next, the level of polarization towards Brexit related policy topics developed between 2015 and 2019 as seen in the official manifestos of the relevant political parties was researched. And lastly, the presence and power of UKIP was also investigated as in indicator of polarization, as extremist parties are a symptom of polarized political systems. It was hypothesized that the level of polarization has increased between 2015 and 2019 between the Conservatives and the other relevant political parties and that UKIP held significant power, despite having no or only 1 seat in parliament, due to their voter share. The literature review showed that it is reasonable to assume higher levels of polarization had a negative effect on the ability to negotiate a withdrawal agreement. In order to investigate the level of polarization, the manifestos were scored by an adjusted RILE-Index which focussed on Brexit related policy categories, whereas the standard RILE-index is purely left-right. Furthermore, the voter share and seat share of the political parties was researched. The results were that polarization did not increase between 2015 and 2017, but significantly increased between 2017 and 2019. The voter share showed that UKIP had significant political power in the wake of the 2015 election despite only having one seat in parliament, because they held almost 13% of the popular vote, enough to greatly swing an election in any direction.In conclusion, when it comes to cooperation and forming coalitions between parties polarization is expected to have a negative effect on their bargaining ability as they become more reluctant to work together. It also appears to have effects internally as the Conservatives held the majority for most of the time, but were not able to get a withdrawal agreement ratified for a long time. There is also support for the notion that UKIP had a relevant polarizing effect as their effective political power was substantial through their share of the popular vote in 2015. This is further supported by the Conservatives ‘caving’ to allow a Brexit referendum and the replacement of Theresa May with Boris Johnson. A former remainer who did not push for a hard Brexit or a Brexit at all costs, was replaced by a Prime-Minster who wanted to get Brexit done at all costs. No deal if necessary if the deadline pass. This was much more in line with certain factions in the Conservative Party who were close with UKIP.

(3)

2

List of Figures

Figure 1: Top five concerns in the United Kingdom ... 6

Figure 2: Patterns of interactions (Schmitt, 2016) ... 10

Figure 3: Single biggest issue 2010-2021 for Britain? ... 18

Figure 4: Most important issue facing Britain today? ... 25

Figure 5: Adjusted RILE-index ... 28

Figure 6: Conservative party RILE comparison ... 29

Figure 7: SNP RILE comparison ... 29

List of Tables

Table 1: The standard RILE-index ... 19

Table 2: Categories kept from standard RILE-index ... 22

Table 3: Categories removed from standard RILE-index ... 22

Table 4: Categories added to adjusted RILE-index ... 23

Table 5: The Adjusted RILE-index ... 24

Table 6: Number of seats in the House of Commons... 30

Table 7: Percentage of popular votes received ... 30

Table 8: Percentage of seats in House of Commons ... 31

Table 9: Difference between percentage of votes and seats in the House of Commons ... 31

(6)

5

1. Introduction 1.1. General Context

One of the main objectives of the European Union (EU) is to increase the economic and political cohesion between its member states (Art. 174 TFEU). The EU believes that unity is power in international negotiations and the more the EU can negotiate as ‘one’ the stronger its position is. On June the 23th 2016 the United Kingdom (UK) voted to leave the EU, on December 24^th 2020 the House of Commons ratified the withdrawal agreement and the United Kingdom has officially left the European Union it on the 31^st of January 2020. As a result of the United Kingdom leaving the EU the relationship between the EU and the UK will had to be renegotiated. These negotiations and how the UK should approach them have been a controversial subject. Although there is a general agreement in the United Kingdom’s Parliament that they should execute Brexit, there is no such agreement when it comes to exactly how it should be done. The options vary from a no deal hard-Brexit to a more Norwegian model where the UK would follow most of the EU regulations, but have no official say in it. The topic of the thesis is centred around on how the internal political cohesiveness, or lack thereof, in the UK could have affected the United Kingdom’s bargaining position.

1.2. Research Questions

There has been a lot of divisiveness on the House of Commons regarding the withdrawal agreement, both between and within parties. In the lead-up to the 2015 election several Labour figureheads advocated against Brexit, but the newly elected leader Jeremy Corbyn stated that he voted against joining the European Union and would do so again. The Conservatives were divided in a similar manner. Prime-minister at the time David Cameron agreed with holding a referendum, but resigned after the British public voted for Brexit. Signaling that although he respects the wishes of the people, he felt he was not the right captain to lead towards Brexit. The Conservative Party was pro-Brexit, but even though they had a majority after the 2015 election they could not get a withdrawal agreement through parliament. They lacked the support from all the Conservative members of parliament and failed to gain the support of other parties.

This thesis will focus upon the effect political cohesion could have had on the bargaining ability of the Conservative Party with the other political parties in the United Kingdom. The general question the thesis will attempt to answer is: ‘How has the level of internal cohesion regarding Brexit related policy matters in the House of Commons developed between 2015-2019 and could this have affected their bargaining power in the Brexit negotiations’ The goal is not to work towards a policy solution, but to shed light on the level of internal political cohesiveness in the UK and how it could have affected their bargaining position and ability to make a deal. As variable of internal cohesion this thesis will use polarization in the policies of the relevant political parties in the House of Commons as expressed in their official manifestos. The hypothesis is that polarization between the parties in the House of Commons of the United Kingdom increased and that it could have had a negative effect on the ability for the House of Commons to reach an agreement on the withdrawal agreement for leaving the European Union. If this hypothesis is supported by the data it is important to note that this would only confirm a correlation between polarization and the difficulties the United Kingdom’s parliament has had on ratifying an agreement. Polarization doesn’t mean no cooperation is it only an indicator and not a guarantee of an outcome.

The basis of the thesis is that political cohesion affects the bargaining power of a party in international negotiations, but how exactly it is affected with need to be investigated. This leads to the following sub question: ‘’How can political cohesion affect the bargaining power of a party in negotiations?’’ This question will be answered through a literature review.

(7)

6 To determine the impact that polarization can have on the UK’s bargaining position, the level of polarization in the UK will need to be determined. Leading to the following sub-questions: ‘’How has the level of polarization developed in the United Kingdom’s parliament in regards to Brexit between 2015 and 2019?’’ To achieve this, the RILE-index as designed by MARPOR and its predecessors will be adjusted. The standard RILE-index focusses on right-left politics and aims to place the political parties on a scale from right- to left leaning by analysing their manifestos and categorizing their statements.

The policy estimates deriving from the analysis of party manifestos can be considered as the most objective source of information (Marks, Hooghe, Steenbergen, & Bakker, 2007, p. 27) as manifestos

‘chart the party’s position at each election on the basis of its own authoritative policy pronouncement’ (Budge I. , 2002). In this thesis the RILE-index will be adjusted and categories that are not related to Brexit will be removed, while others that are will be added. The adjusted RILE- index will more clearly show if there is any polarization between the relevant political policies’

regarding Brexit relevant topics. This will not necessarily correlate with the ‘hardness’ of the Brexit desired by the political parties. A party can be more left than another party, but wish to achieve those goals through or with the European Union. While the other party wishes to achieve theirs despite or without the European Union.

An indicator of polarization is the existence of fringe extremist parties (Sartori, 1976).

However, a problem that arises is the question how serious should those fringe parties be taken into the overall view of polarisation among the political actors. Interactions and conflict between less important and relevant actors do not carry the same weight as when those occur between the larger actors. There are two indicators (Ezrow & Xezonakis, 2011) (Alvarez & Nagler, 2004) that imply the importance of those parties. Firstly, their seat share in the UK’s parliament and secondly, their political relevance. The United Kingdom has such a party in the UKIP, but their political power can be hard to determine. Their presence indicates at least some level of polarization present within the system, but if they held a significant level of political power then that would be a clearer indication than mere existence. Furthermore, if they held significant political power then that would shed more light on the political movements of their ideologically most closely related parties.

Figure 1: Top five concerns in the United Kingdom (Ipsos MORI Issues Index: March 2021)

As seen in figures 1 and 3 Brexit went from almost nobody’s top concern to the most important issue for the vast majority of the population of the United Kingdom. The Conservative Party under leadership of Cameron also faced strong internal pressure to hold a Brexit referendum and eventually relented. Yet UKIP held no seats in the House of Commons after the 2010 election and only two after the 2015 elections. How were they seemingly able to exert this pressure on the Conservative Party?

A possible reason for this lies in the way the United Kingdom elects its members of parliament.

There can be a large discrepancy between the percentage of votes received nationally and the

(8)

7 percentage of seats in parliament. The parliament of the United Kingdom consisted of 650 parliamentary constituencies during the 2010, 2015, 2017 and 2019 elections. Each constituency elects a single member of parliament into the House of Commons using a first past the post system.

Therefore to analyze the (potential) political power of UKIP this thesis will look into the percentage of votes received and percentage of seats in parliament of the national elections of the United Kingdom. If UKIP holds a significant percentage of the votes then they could be a real threat to the established political parties despite having no or almost no seats in the House of Commons. This would also indicate a higher level of power for an extremist party and be further indication for increased polarization. The hypothesis to be tested is that UKIP has far more power than their seats in the House of Commons indicates because of the proportional number of votes received by UKIP.

Leading to the sub-question: ‘Does the percentage of proportional votes received by UKIP support the notion that they had a relevant polarizing effect on the other political parties?’.

1.3. Academic and Practical Relevance

One of the pillars of the EU is ‘unity is power’, and Prime-Minister Theresa May has also expressed this sentiment for the UK (Hall, 2018). It is crucial that both the EU and the UK develop a clear strategy for the future and Brexit is where both the EU’s and the UK’s internal cohesion is put to the test (European Commission, 2014). Legislative party discipline and internal political cohesion are important subjects when analyzing political systems. Although generally speaking it is assumed that the political parties are cohesive and act as a unified collective with, reasonably, clear goals. The United Kingdom does not appear to have unified parties, let alone a unified parliament to support the government. Without a cohesive governmental party and parliament government survival in parliamentary systems is difficult and unstable. Something which can be seen in the United Kingdom in the frequent leadership challenges in both the Conservative Party and the government itself.

There is currently no research on the internal cohesion of the UK regarding the Brexit negotiations and the effect a lack of it can have. Was the UK able to form a coherent, targeted and dominant negotiation position in the Brexit talks? If not, what effect could the internal cohesion have had? And did the UK parliament and other institutions put their trust in the Prime-Minister and his or her cabinet to navigate the UK through Brexit? The effects a drastic like Brexit move can have on the internal cohesion of a state, not to mention one that consists of several nations, in modern times has not been researched yet. The EU is also in a unique position, since this is the first time a state is actually leaving the EU. It is interesting to take a look at the UK and what kind of negotiating partner the EU has and will be dealing with since that could play a (big) role in how the UK acts after Brexit.

The EU’s history as a multilateral and bilateral participant is mixed. Multilaterally it is seen as a positive, but oftentimes ineffective player, while bilaterally it seems to lack the coherence to form a converging dominant strategy (European Commission, 2014).

2. Theoretical framework

Political cohesion and party discipline are important factors in the study of political science and have been extensively researched (Owens J. E., 2003). They form a core aspect of parliamentary systems through legislative cohesion and its impact on the stability of the government (Boucek, 2012;

Saalfeld, 2009). Furthermore, the cohesiveness of a party also has an influence on the bargaining power during the coalition making process after an election (Giannetti and Benoit, 2009; Pedersen, 2010). This thesis will look at the key issues that will have to be negotiated in the Brexit deal and determine which ones have a centripetal effect and increase political cohesion and which ones have a centrifugal effect and thereby decrease political cohesion. However, political cohesion is broad and abstract concept with multiple dimensions and can be defined by a plethora of empirical methods (Close, 2016a). This thesis will look into polarization specifically as an indicator for political cohesion.

(9)

8

2.1. Polarization

Polarization was chosen because it is one of the more established and used indicators when researching the outcome of a political system (Dalton, 2008) (Curini & Hino, 2012) (Sartori, 1976).

Furthermore, the discussion around what Brexit the UK should aim is often held along the soft Brexit or hard Brexit line. A divide such as that is similar to the research done into the polarization of left and right leaning political parties (Kam, Indridason, & Bianco, 2017) (Mölder, The validity of the RILE left-right index as a measure of party policy, 2016) (Schmitt, How to Measure Idealogical Polarization in Party Systems, 2016). This would provide the opportunity to go beyond a more basic measure of only categorizing how many actors disagree. The traditional polarization research already places political parties on a scale from left to right and this can be as a basis for the similar soft Brexit – hard Brexit line and allow the opportunity to be able to place political actors on a scale of disagreement surrounding the Brexit approach. However, despite the abundant use of polarization in research it is often accompanied by a vague definition of polarization. It is generally described as the ideological differences between actors in a particular environment (Schmitt, How to Measure Idealogical Polarization in Party Systems, 2016). The different approaches to polarization will be outlined below.

Sartori’s concept of polarization is oftentimes referred to when discussing polarization (Sigelman & Yough, 1978) (Knutsen, 1998) (Pelizzo & Babones, 2007). However, this initial definition is still rather short, Sartori (1976, p. 126) defines polarization as follows: ‘’The term is used first to denote an ideological distance, that is, the overall spread of the ideological spectrum of any given polity […]”. Sartori (1976) continues later on with the argument there is polarization when there is a

‘ideological distance’ between parties, rather than ideological proximity. For this ideological distance to exist there needs to be a dominant ideological dimension present in the party system. Sartori (1976) argues that there needs to be some sort of ideological structure for there to be a rise of strong ideologically differentiating parties to polarize the political system. If such a dominant ideological dimension is absent from a party system then the parties will segment instead of polarize.

Even when researchers give their own concept of polarization, they oftentimes only offer slight variations, such as Dalton (2008) who defines it as ‘’the degree of ideological differentiation among political parties in a system.’’. Similar approaches can be found in the works of Klingemann (2005), Han (2015) and Pardos-Prado & Dinas (2010). The core of polarization, in this thesis, is therefore build on the cumulative ideological differences between the political actors in the United Kingdom’s parliament. For the purpose of this thesis the ideological differences will be based on the approaches the political actors want to take in regards to Brexit. Where the extremes are not left and right, but no Brexit and hard Brexit. Several authors also include the electoral success of extremist parties as an indicator for high levels of polarization (Pelizzo & Babones, 2007) (Warwick, 1992).

They argue that a high level of polarization in an environment also suggests the presence and relative electoral success of extremist parties. The concept was initially coined by Sartori (1976) who also argued those parties were anti-system and had anti-constitutional characteristics. However, since then the idea that anti-system parties also have to be anti-constitutional has been reassessed (Capoccia, 2002). Capoccia (2002) writes that when a party is positioneted a significant distance away from other parties on the scale then they can still be caterorised as anti-system, but the ‘system’

they’re against refers to the ideological system of the more established parties.

Both the distribution of the parties and the presence of extremist parties depend on the sub- concept of ideology, with most studies using the left-right dimension as the underlying basis (Schmitt, How to Measure Idealogical Polarization in Party Systems, 2016). However, the operationalization of the left-right dimension varies. One option is to create a specific set of issues and apply them to all political systems equally, regardless of the specific context of a political system, an example of this is the Downs’ (1957) economic left-right dimension. Another approach is looking at the right-left spectrum as a ‘’super issue’’ (Schmitt, How to Measure Idealogical Polarization in Party Systems, 2016) which encompasses all the revelant issue and policy positions between the

(10)

9 parties in a specific party competition. The thought behind the second approach is that the behaviour, priorities and issue stances of the parties in a particular party system can change over time, or change for a particulaier situation, which in this thesis is Brexit. This approach does lessen the significance of comparison between countries since they and the research will focus on different things within that political system at a particular time, which is why such comparative research, done by MARPOR for example, uses the first approach. Furthermore, there is also a discussion on the amount of dimensions used in these left-right scales (Schmitt, How to Measure Idealogical Polarization in Party Systems, 2016). One dimensionality is regularly criticized for being to basic to properly represent the complex nature of party systems. However, more dimensions can also make the research too complex to be able to extract any meaningfull conclusions. In the end, most researchers have chosen to approach polarization one dimensionally this thesis will follow in that example.

2.1.1. Interaction between actors

Polarization in a political system leads to deep (social) cleavages, low consensus among the political parties and the political system’s legitimacy will be questioned (Sartori, 1976). Vegetti (2014) expands on this and argues that these developments do not imply just policy dispersion, but actual political conflict where the overall conflict is more important that specific clashes. This is because the parties believe non-cooperative tactics have a more promising future, which makes coordinating and distributing policy and resources harder and leads to conflicts (Rodrik, 1999). Bartolini (1999) describes the four escalating types of actor interaction as: cooperation, negotiation, contest and conflict. Cooperation assumes a common goal and relatively similar method of achieving that goal between actors and they are willing to work together to achieve that goal. Negotiation is similar to cooperation in that the basis behind it is solidarity, a desire to cooperate. However, with negotiation the actors do have different goals, but they are willing suppress their self-interest to a certain degree to be able to achieve a new shared goal. Once agreement has been reached between the actors, then the relationship between the actors can be seen as cooperation rather than negotiation.

Contrarily, if no agreement has been reached then the relationship could escalate towards the contest type. Contest is the opposite compared to negotiation in its relations with cooperation. In the contest type there is a common goal between parties, but they are not willing to suppress they self-interests. The actors will actively attempt to achieve their goals, regardless of the interests of others. In the conflict stage there is no common goal between the actors and furthermore, the actors will not only strife for their own goal, but they will actively attempt to damage and undermine one another.

As the level of polarization rises, but still within moderate levels, the more the political system moves towards more contests. The actors still generally follow the rules and structure of the political system, however their ideological ideas differ too much to cooperate with each other (Bartolini, 1999). However, if the level of polarization keeps rising then the party system consensus will lower further and as a result of this more conflicts will arise between the actors. The ideological differences between the actors will start make it impossible to find any agreement between the actors. However, it is important to note that ideological distance only encourages certain behavioral types and does not predict them, i.e. two actors having similar ideologies does not necessarily mean that they cooperate with each other (Franzmann, 2011). However, having similar ideologies is a prerequisite for cooperation as is having significant ideological differences if there is to be conflict.

Negotiate and contest are similar to cooperate in this regard, but they allow for larger ideological differences than cooperating. Furthermore, cooperate, negotiate and contest are characterized by the fact that the actors follow the unwritten rules of the political system and are dueling with each other based on the issue at hand instead of disagreeing with the other actor solely because of the actor (Bartolini, 1999) (Sartori, 1976).

(11)

10

Figure 2: Patterns of interactions (Schmitt, 2016)

2.1.2. Extremist parties

Polarization is a complicated notion to measure in a political system, at its core the result should reflect the distribution of the political actors over an ideological dimension. However, extremists parties are both an outlier and a phenomenon of polarization within a political environmental, as will be discussed in this chapter. This thesis will measure the polarization in the UK’s parliament regarding Brexit on a one dimensional scale, namely no Brexit versus hard Brexit. This means looking at both the polarization between parties and the polarization within parties. If the political actors are congregated around each other, then that would point towards centripetal forces that are encouraging the actors to move towards each other. However, it is important to note that it does not necessarily have to be in the centre of the scale. Following Sartori’s (1976) view it is not necessarily about actors around the centre, but about the ideological proximity of the actors in relation to one another. Polarization in that view would be when centrifugal forces are pushing the actors towards the extremes, away from each other. However, another indicator of polarization is the existence of fringe extremist parties which can be present in both situations. These parties are the effect of centrifugal forces in the political system and can reinforce that effect as well. However, a problem that arises is the question how serious should those fringe parties be taken into the overall view of polarisation among the political actors. Interactions and conflict between less important and relevant actors do not carry the same weight as when those occur between the larger actors. There are two indicators (Ezrow & Xezonakis, 2011) (Alvarez & Nagler, 2004) that imply the importance of those parties. Firstly, their seat share in the UK’s parliament and secondly, their political relevance. Seat share in favour of vote share is important due to the election methods in the United Kingdom and because this thesis looks at the polarisation in the parliament. An actor could have, for example, received a number of votes, but not have a presence within the parliament itself. Another example is that party A could have received more overall votes than party B across districts, but party B’s votes were more concentrated in particular districts resulting in one or more seats in parliament.

Meanwhile Party A has less or even no seats in parliament. Party A would have a larger share of the votes cast, but they would have significantly lower direct political power in the current parliament from their lack of seats.

However, political relevance can compensate for a lack of seats to a certain degree. Political relevance is a more subjective term and looks at what a party can and has achieved with their seats.

A party with more seats in parliament does not necessarily have more political significance in all circumstances. An example of this in the United Kingdom is the DUP. They held a relatively low amount of seats in parliament under May’s government, but their support is what provided the government with a majority in parliament. Looking at just their share of seats undervalues their political significance and the centrifugal effects they can have on the government and the opposition by pulling the government to a certain direction. Another party that had a similar, but far larger

(12)

11 effect is the populist party UKIP. UKIP starting gaining importance in the early 2010s as the far right anti-EU membership party, spearheaded by long-term member and leader Nigel Farage, and received the ‘major party status’ from Ofcom in the 2014 local elections (Deacon & Wring, 2016). Although they did not take over control of a council, they did manage to become the British party with the largest number of votes in the elections of the European Parliament ever. This was a big victory as it was the first time in over a hundred years that neither the Labour Party nor the Conservative party had received the most votes in a nation-wide election (Goodwin & Milazzo, 2015, p. 104). The year also marked the election in which UKIP gained its first Member of Parliament when a conservative defector won the election of the seat of Clacton after resigning and triggering a by-election. 2014 Made UKIP into a ‘big party’ and a household name in the United Kingdom (Goodwin & Milazzo, 2015, p. 4). However, even though UKIP made large gains in even traditional Labour areas and they received 12.6% of the total votes in the general election of 2015, thereby becoming the third largest party in number of votes, they only won a single seat in parliament. This single seat made them the tenth largest party is Parliament and was disproportional to the political relevance they had gained in the political power sphere of the United Kingdom. The conservatives saw the danger UKIP could pose if allowed to continue on its current path and to counter them David Cameron promised to hold a referendum to allow the people of the United-Kingdom to vote on the continued membership of the EU, to counter a party with only a single Member of Parliament (Goodwin & Milazzo, 2015, p. 126) (Reed, 2016). After the referendum had been ‘won’ by UKJP it entered a quick decline and Nigel Farage quit as leader, which is often a hard blow to a populist party due to the direct relationship voters often develop with the charismatic leader (Barber, 2019). It is however a strong indication that the only purpose UKIP, and a large part of their base, wanted to achieve was Brexit and when the Conservatives aligned themselves with this purpose after the 2015 elections the party lost a lot of power. In the aftermath the party wanted to return to its more libertarian roots instead of just being a party focused on Brexit. However in the local election of 2018 the party lost around two out of every three seats it had won in 2014, whilst not winning a single seat in parliament in the general election of 2017 and 2019. Their political relevance dropped even sharper as they were no longer able to leverage the push for Brexit against the Conservative party, because the Conservatives had leaned into it heavily and pushed for Brexit themselves on similar terms.

The political relevance of UKIP might have dropped, but the political relevance of its former leader Nigel Farage had not. In the aftermath of the referendum and in the run-up to the 2019 general election he had found a new battleground to fight upon and a new party to do it with, namely the terms of the Brexit and the Brexit Party. The Brexit party is another anti-EU membership populist party in the United Kingdom (Zulianello, 2020), led by Nigel Farage, founded in November 2018 and became active in January 2019. The parties’ main target is for the United-Kingdom to leave the EU with or without a new trade agreement, reverting to standard World Trade Organisation trading rules if needed. Similarly to UKIP the Brexit Party did very well in the elections of the European Parliament and became the party with the most seats. In the run-up to the 2019 general election in the United Kingdom, Nigel Farage offered a deal to the Conservative party for an electoral pact (Kirka, 2019). Even though the conservatives refused the deal, the Brexit Party decided not to field candidates in the constituencies where the conservatives had won in the previous election. This cooperative stance towards the Conservative Party is a turnaround from the more antagonistic approach from UKIP, even though in terms of goals the Brexit Party is very similar to UKIP. This indicates the effect UKIP, a party with never more than a single seat in parliament, has had on the Conservative Party and British politics in general. The Conservative Party moved to the right enough and towards a strong Brexit stance enough that the Brexit Party sees far less need to try and take votes from the conservatives. It also anchors the Conservative Party in at least their current position as a softer stance on Brexit could see a rise in Brexit Party support from their voters.

DUP, UKIP and the Brexit party are all three parties whose political relevance far exceeded their seats in parliament. However, how they wielded that power differentiated substantially. The DUP

(13)

12 mainly focused on Northern-Ireland and its border with Ireland. UKIP and the Brexit party were far more polarizing entities in British (identity) politics. The relatively cooperative stance from the Brexit Party towards the Conservative Party shows that the Conservative Party has moved to the right and a stronger Brexit stance significantly. To ascribe that move entirely to UKIP would be premature, but the notion that UKIP has had a polarizing influence on the Conservative Party is not unreasonable.

This shows that a party can have significant political relevance in the absence of direct political power if the (extremist) party can find a charismatic leader with a strong leverage point.

2.2. Literature Review: Bargaining power

The question that this literature review attempt to answer is ’How can polarization affect the bargaining power of a party in international negotiations?’. To answer this question, firstly the type of bargaining power will be established and then the effect polarization can have on the bargaining strength of a party in such a negotiation. It builds upon the effects of polarization upon actor interactions as shown in figure 2

The effect of coalitions on bargaining power is taken into account because of a development in the UK’s political parties, especially in the conservatives. In the wake of the Brexit referendum factions within the political parties have increased the pressure on the party at large in regards to the Brexit negotiations. Many parties had internal strife in how to approach the Brexit negotiations, with internal factions pushing for hard Brexit at all costs compared to more mild stances of the party at large or vice versa. This resulted in party leadership having to spend time convincing not only their constituents into believing and other parties into cooperating, but also having to negotiate and convince, or sometimes contest, with factions in their own party. Prime-Minister May failed to appease the internal factions within the conservative faction and this resulted in a leadership challenge in which she failed to appeal to her constituents, which allowed Boris Johnson to become Prime-Minister. This challenge was build on the much harder Brexit stance Boris Johnson was aiming to take.

There are always factions within the party, but increasing polarization will increase the presence and effect of those factions. This changes the party more to a coalition of factions than a unitary actor. The time and effort on building and maintaining internal cohesion among the internal political actors increases. This lessens the power and position of the party and makes it more difficult to act (Wasserman & Faust, 1994). It will also lessen the chances of a coalition of parties working together to achieve their desired Brexit goals in places where those goal align. Increased polarization will furthermore push the preferred outcome of the other parties further away from the Conservatives, which will make it harder for the Conservative government to create a majority coalition with the other parties. Even if their Brexit goals should align cooperation could be impossible if the parties are too much in conflict with one another or if their supporter base does not want to see ‘their’ party helping the ‘enemy’.

2.2.1. Type

The two main bargaining types that can be ascribed to the UK-EU negotiations are distributive bargaining, where one party benefits at the others’ expense. And integrative bargaining whereby both sides are more aimed at a win-win solution. Both parties have an agenda and they try to reach a consensus that benefits both parties. Since the Brexit negotiations are some of the biggest negotiations ever held with some very specific negotiations areas that are dealt with separately, it is possible that for some areas a more distributive bargaining is held, where the other parties’ consent is bought with side deals. However, for the purpose of this thesis is will be assumed that the entire negotiations will fall under one type, to let the scope of the thesis not grow too large. The attitude of

(14)

13

‘winning’ the negotiations are seen as a method of domestic approval and to win domestic political support among the political actors of their party and the populace.

2.2.2. Power and motivation

The network bargaining model takes the view that power is what drives negotiations, specifically, that actors use the power available to them to support their preferences towards the issues outcome (Boddewyn & Brewer, 1994) (Gourevitch, 1999). The influence an actor can use during bargaining comes from two factors (Brewer, 1992). Firstly, the actors base of power. This refers to the power the actor has over the other actor(s) in the negotiations. This can be resources, ability to pass and enforce laws, access and influence over other/external actors, capital or that the other actor has a political or economic dependence on it. Secondly, the willingness and motivation of the actor to use and apply this power in the negotiations. The ability to convert ones resources to influence another actors is the definition of power in negotiations (Blalock, 1989) (Burt, 1977). The influencing can either be favorable or unfavorable. This power can be present by executing it currently, or by having the ability to do so when the actor desires it. This definition of power has also been referred to as

‘actual power’ by Kobrin (1987) or ‘effective power’ by Keohane and Nye (1977). There is also an additional layer, described as ‘potential power’ (Burt, 1977) (Gourevitch, 1999). However, the ability to use this potential power in the negotiations may never be achieved due to a number of reasons (Nebus & Rufin, 2010). The potential power is unacknowledged in the real negotiations, unless it is apparent that the actor can readily use the power, it will not be taken into account. An actor’s actual power is evident, potential power is doubtful or perhaps even unobservable.

The degree to which an actor is flexing its influence towards an issue is determined by the actors’

motivation to apply the power it has available (Blalock, 1989) (Mahon, Heugens, & Lamertz, 2004) (Slaughter, 1993). An actors’ motivation is in turn determined on the actors’ stake in the outcome of the issue. An actor will not devote many resources to issues it does not care about and vice versa.

However, if an actor is seeing that an issue it cares about is already heading towards its preferred outcome, regardless of actions the actor takes, then it might not be motivated to apply its resources even though the actor cares about the outcome (Nebus & Rufin, 2010). As a result the motivation factor is issue specific. Regardless of the reasons that push an actor to (not) act, the actors’ power in the bargaining outcome is only relevant to the degree is it motivated to apply it. Motivation can only decrease the potential power an actor can apply, it cannot increase it and therefore it is considered a moderating factor on the bargaining influence of an actor (Nebus & Rufin, 2010).

2.2.3. Power and Coalitions

Actors do not always act alone, in many situations an actor will work together with one or more other actors towards a common goal, such as a blocking or pushing a policy position. Actors with identical, or similar enough, policy position can choose to form a coalition, thereby acting as one actor. When the power of the coalition rises above the power of its most powerful member then the most powerful actor is no longer the individual, but the coalition (Nebus & Rufin, 2010). This is despite the fact that that actor would still be the most powerful member of the coalition. Because of this it is important that the policies of the coalition members are aligned with one another, as multiple actors with similar goal can form a coalition to gather their political power.

When actors form a coalition it is safe to assume it is with the goal of increasing their power beyond the level of the individual actors. However, the scaling of this power is not calculated by additively adding the political power of the actors (Nebus & Rufin, 2010). Instead, the sum of the political power of the separate actors is the minimum base power of the coalition. This is the case in situations where the parties to the coalition bring disparate resources (Nebus & Rufin, 2010) to the table and when they bring complementary resources, so long as the complementary resources are

(15)

14 the bases of power in the coalition (Milgrom & Roberts, 1995) (Nebus & Rufin, 2010). An example of disparate resources is one party supplying monetary funds, while another party has an advantageous bargaining position or superior negotiating skill. Examples of complementary resources are seats in parliament. The individual actors might not have the majority in parliament, but if the coalition could achieve that, or could reach the size of a blocking coalition for certain legislation (Tsebelis, 2002).

Coalitions also introduce some inefficiency into the equation in the form of two primary issues, which Hechter (1987) refers to as control and dependence. Control refers to the need for internal control of the contributions of the coalition partners. Members may attempt to coast on the contributions of others, only taking, but not adding anything. This effect is strengthened when the coalition members have been in conflict with each other in the past, or if they expect to be in conflict with each other again in the future. If an actor believes the other members will devote enough resources to achieving the desired goal than it can choose to conserve its own resources for those future clashes, despite the importance of the issue at hand. In order to combat this, the coalition will need to monitor and incentivize the coalition members to ensure compliance. This usually means that some of the coalitions’ external power is lost as it is directed inwards (Hechter, 1987).

Dependence refers to a situation where not all the parties of the coalition are equally motivated to participate in a particular issue, even though they share the coalitions’ preference regarding the outcome (Hechter, 1987). Furthermore, even in a situation where the coalitions’ actors are all fulfilling their responsibilities and are equally motivated towards their objective, using the power they have is more difficult for them than for single actors. The coordination of multiple actors in a coalition, especially without central authority, creates more inefficiency than even more chaotic single actors. This effect is strengthened in intersectoral coalitions compared to intrasectoral ones. If the coalition is made up of intersectoral actors than the cohesion can be improved by having similar types of actors involved (Nebus & Rufin, 2010).

For many issues the political actors forming a coalition, or can be approach for one, are obvious, from their publicly stated policy or their behaviour or both. However, for some issues it is not clear what actors are in the coalition, or can be approached to join. The unitary actors will want to join a coalition that maximizes their political power, but still maintains their preferred issue outcome. To this end they generally willing to cede some influence, or compromise on their preferred outcome, if it adds to the coalition’s power, especially if it prevents unwelcome outcomes being pursuit by opposing parties. This mechanic is referred to as preference alignment; it refers to the degree of correlation between the preferred outcomes of multiple actors (Nebus & Rufin, 2010). The lower the alignment, the lower the chance that actors are willing to enter in a coalition with one another.

2.2.4. Polarization and bargaining

Polarization can have a weakening or a strengthening effect on coalitions. As the level of polarization rises the system moves towards more contests between actors and parties that are polarizing will become too ideologically different from ‘the other side’ so cooperate much, or at all of polarization increases enough, with each other (Bartolini, 1999). In multiparty systems such as the UK this can also encourage new coalitions if multiple parties are polarizing to a similar direction, due to needing each other to achieve their new similar goals. However, two actors having similar ideologies does not necessarily mean that they will cooperate in a heavily polarized system (Franzmann, 2011). Lesser ideological distance is merely a prerequisite, it only encourages cooperation, it does not predict it.

Polarization also introduces a new factor in cooperation and negotiation. In non-polarized systems actors will cooperate and contest based on the issue at hand, however with a high enough level of polarization actors will start contesting, or refusing to cooperate, based on the actors involved instead (Bartolini, 1999) (Sartori, 1976).

An increasing level of polarization in the UK political system would therefore make cooperation between parties harder and more unlikely. Furthermore, the rise of factions within

(16)

15 parties also moves the parties more towards a collection of factions working together than a unitary party. This is always somewhat the case, but it has increased in the last couple years as the Brexit situation evolved. Polarization towards Brexit within parties would have a similar effect as between parties. The power of the political actors experiencing polarization within its ranks would decrease and it would make it harder for the actors to act and exert power in the political system. An example of this is the snap election called by former Prime-Minister May. The narrow majority held by the conservatives would theoretically be enough to enact the Brexit the conservatives want. The reason for the election was that that May wanted to increase their majority to make it easier to enact Brexit.

This is a signal that cooperation between the parties was becoming harder. May and/or the conservatives as a whole felt that other parties could not be relied upon to support the Brexit the conservatives wanted. Or the conservatives were not willing to concede on certain matters to allow for cooperation. Either way one or both sides were not willing to work together on the issue. Now, it is not certain that this is due to increasing polarization. As mentioned before polarization is only a prerequisite for less or more cooperation and there can be many other reasons for the lack of cooperation between the parties. However, it is interesting that both Labour and the Conservatives were strongly, although not unanimously, in favour of leaving and together they controlled a vast majority of the seats in parliament yet they refused to work together and the Conservatives favoured another general election over cooperation. With the risk that the election could cost them the majority they had on their own, which it did. The parties had the power to enact Brexit, but lacked the motivation to do so under the circumstances. The Conservatives were willing to risk their majority, but why? An argument could be made that they were convinced they could increase their majority, possibly by taking the UKIP votes through strongly pushing for Brexit. Another factor could have been that May did not believe she had a majority because of fractions between the factions within the conservative party and she believed that the Brexit supporters from the other parties would not follow her in her Brexit Proposal. Or it could have been both factors working together. In any case, it would be interesting to see whether the level of polarization seen in the RILE and adjusted RILE shows an increase in polarization over the years between the parties that could explain their refusal to cooperate.

Since when the level of polarization rises, but is still within moderate levels, the more the political system moves towards contests and conflicts. During contests the actors still generally follow the rules and structure of the political system; however their ideological ideas differ too much to cooperate with each other (Bartolini, 1999). However, if the level of polarization keeps rising then the party system consensus will lower further and as a result of this more conflicts will arise between the actors. The ideological differences between the actors will start make it impossible to find any cooperation between the actors, which was observed between the political parties in the deal making process. Only ‘their’ deal was acceptable to the parties. However, it is important to note that ideological distance only encourages certain behavioral types and does not predict them, i.e. two actors having similar ideologies does not necessarily mean that they cooperate with each other and vice versa (Franzmann, 2011). However, having similar ideologies is a prerequisite for cooperation as is having significant ideological differences if there is to be conflict. Negotiate and contest are similar to cooperate in this regard, but they allow for larger ideological differences than cooperating.

Furthermore, cooperate, negotiate and contest are characterized by the fact that the actors follow the unwritten rules of the political system and are dueling with each other based on the issue at hand instead of disagreeing with the other actor solely because of the actor (Bartolini, 1999) (Sartori, 1976). When the actors are in conflict they will refuse to cooperate based on the fact that the cooperation requires working with the party they are in conflict with.

This attitude can regularly be seen towards extremist right wing parties. An example is the general refusal in the Netherlands to cooperate with the PVV to form a government, despite the fact that the party is among the biggest. However, this cannot be laid solely on the effects of polarization and political attitude, extremist right wing parties tend to be populist in nature and that can make

(17)

16 those parties unpredictable and unstable. The other parties might be in conflict enough to only avoid long-term cooperation with parties such as that. The others parties are willing to work together on shorter term problems, on a case by case basis, where their preferences align almost perfectly, although it tends to be the opposition who joins the PVV against the government, instead of governmental parties choosing to work with them. Which is not unexpected since the PVV generally holds around one third of the seats in parliament not claimed by the governmental parties. It could be argued that they are forced to cooperate if they which to accomplish something against the government when those parties are united.

Motivation to work with another party can therefore also be seen as a balancing act between their perceived importance of the issue and desired goal, and their reluctance to cooperate with the other actor. The Brexit negotiations scores high on importance. The negotiations will have strong and long lasting consequences on the UK, they are perhaps the most important negotiation the UK will hold for a long time. The parties’ reluctance to cooperate must be very high to stop them from cooperation. Especially since Brexit adds another dimension, namely, that negotiating no deal also results in a ‘deal’. The relations between the EU and UK would revert to WTO rules and many sudden changes to the relations between the EU and UK as a result. The parties in the UK have generally spoken out against a ‘no deal’ situation, which should push them towards cooperating, but that cooperation has not occurred. This thesis will not go in depth as to the entire reason why this cooperation has not occurred, but will focus on polarization and the role it might have played in this refusal to cooperate. It will not explain the entire refusal as it is not the only factor and it is only an indicator and not a guarantee of an outcome.

3. Research Methodology

3.1. Adjusted RILE and Proportional vote

The core of the thesis is going to be focused on the polarization in the UK’s parliament in regards to their policy preferences towards topics related to Brexit. The Research question is causal in nature and will be answered with qualitative data. To gather the data a case study on the UK will be done.

The manifesto’s of the most relevant political parties in the UK are processed according to the RILE method as set out by the MARPOR institution. Normally the objective of a case study is to study something in specific cases and then extrapolate that to the broader population. However, this extrapolation is absent in this thesis. Instead it will look at a single outcome, namely, polarization in the UK’s parliament. Gerring (2006) attempted to distinguish this type of case study and called it a

‘single-outcome study’. The main research question focusses on polarization with a focus on Brexit related topics and therefore the timeframe of the data collection will be the manifestos from the 2015, 2017 and 2019 national elections of the United Kingdom. The level of analysis will be the UK parliament and the units of analysis will be the political parties of the UK parliament.

Another indicator of polarization is the existence of fringe extremist parties; these parties are the effect of centrifugal forces in the political system and can reinforce that effect as well. However, a problem that arises is the question how serious should those fringe parties be taken into the overall view of polarisation among the political actors. To support the hypothesis that UKIP is politically relevant and has had a polarizing effect on the political parties of the United Kingdom their proportional share of the votes will also be investigated. If they received a significant percentage of the votes, despite few or no seats in the House of Commons; that would support the notion that they have had a polarizing effect, provided there was such an effect.

(18)

17 3.1.1. Adjusted RILE

To measure the level of polarization regarding Brexit between the parties in the UK’s parliament the right-left (RILE) scale from the MARPOR (Volkens, Bara, Budge, McDonald, & Klingemann, 2013) will be used and slightly altered to fit the Brexit context. The possibility to determine the general position of political parties on a scale from left to right is one of the more popular applications of the Manifesto Project Dataset. The categories on which the measurements are based were originally theoretically confirmed, but later empirically confirmed with a factor analysis (Laver & Ian, Party Policy and Government Coalitions, 1992). The RILE scale is widely used and has several advantages for the users. Firstly, it is simple to calculate and therefore highly transparent. Secondly, this makes it straightforward to understand and communicate to others. Finally, it can easily be adapted to specific countries and circumstances. Proportional vote

3.1.2. Extremist parties and proportional votes

The presence of extremist parties will not be used as an indicator for polarization in this thesis, only their movement on the RILE scale and their political relevance. What will be included is the percentage of vote shares during the elections. The presence and voters share of an extremist party could be useful in explaining and understanding any movement the party most closely related to it makes on the RILE- and adjusted RILE scale. The sub-question this thesis will attempt to answer is:

‘Does the percentage of proportional votes received by UKIP support the notion that they had a relevant polarizing effect on the other political parties?’ A party with a relatively large voter share compared to their seat share can with a small increase in votes gain a large increase in seats in the parliament during the next election. This power might not directly translate into effective power in parliament, but it will increase it. It will also significantly influence campaigning in the lead up to the next elections. Parties who are ideologically closely related to such a party will be interested in

‘taking’ the many unrepresented votes from the underrepresented party. This could gain them many seats in parliament and would weaken the party most likely to threaten their seats. The underrepresented party would have the same reasons for attempting to take their votes, which would normally mean a movement of policies towards each other. However, extremist parties are generally unwilling to move from their fringe position and it could be argued that their extreme and fringe position is exactly what got them their large voter share. If the voter base is equally unwilling to budge from an extreme position then the only option to coax those voters to your party is to take up, a part of, their extremist views and moving your party further to the extremes. Which is what the Conservatives attempted to do with the Brexit referendum and UKIP and the expectation is that this movement will show strongly in the scales.

3.2. Specification

3.2.1. Relevant Political Parties

Firstly, it must be decided which political parties are relevant for this thesis and will be included in the analysis. This deliberation will include more than the amount of seats in parliament held after a single or multiple elections; although size will still play an important role. The Conservative Party and the Labour party are included because they are by far the largest parties and have formed the Government and opposition for the entire period this thesis encompasses. The Scottish National Party is also included for size, as it has increased in size after the 2015 election significantly becoming by far the third largest party in the House of Commons. The Liberal Democrats the Dublin Unionist Party are included because of size and each has complementary political reasons for relevance. The Liberal Democrats shrunk heavily in size from the 2015 elections onwards, which is an interesting

(19)

18 development as it coincides with the emergence of political relevance of Brexit. Furthermore, they have become the main anti-Brexit voice from the left side of the United Kingdom’s political spectrum.

While Labour as a whole is more divided upon the subject, not only in a Brexit or no Brexit discussion, but in the way Brexit should be achieved. The DUP is interesting because it is the main political party from Northern-Ireland, who are arguably more effected by the Brexit discussions than the other regions. Furthermore, because the Conservative Party lost the majority in the 2017 national elections forcing them to find support from another party in the House of Commons. The DUP and the Conservative Party did not enter in a coalition, but the DUP did agree to back the Conservatives in key matters, although with the exclusion of Northern-Ireland related matters. Lastly, UKIP is included, despite having only 2 seats in parliament in the 2015 election and none in the others. They are included because of their political relevance, especially regarding Brexit.

Figure 3: Single biggest issue 2010-2021 for Britain? (Ipsos MORI Issues Index: March 2021)

As can be seen in figure 3 Brexit is seen as the single biggest issue for the British people between 2015 and early 2020, only being overtaken by the Coronavirus. Otherwise it would be reasonable to assume it would have remained the top concern for longer. UKIP, especially in the 2010-2016 period was the main vocal voice calling for Brexit and ‘taking back control’. They are a single-issue right-wing populist party aimed at achieving Brexit. UKIP, and especially Nigel Farage, have also had a significant media presence throughout the 2015-2019 period and have partnered in these media appearances with significant political actors from other political parties such as Boris Johnson, the current prime- minister, of the Conservative Party. They also received the most votes in the 2014 European Parliamentary election from the United Kingdom.

3.2.2. Data analysis manifesto’s

The basis of the thesis will be the party manifestos from the 2015, 2017 and 2019 general elections of the UK. The manifestos will be processed according to the RILE method, although with some changes to the categorisation as to adjust it for Brexit, as it is the most objective measure of the parties’ stances regarding the various topics that will be analysed in this thesis. The changes made will be expanded upon in chapter 3.3. The overall principle behind the RILE-scale is that left-wing parties will talk more about left wing issues and right wing parties will talk more about right wing issues (Laver & Ian, 1992). Furthermore, their negative attitudes to policy position generally seen as positive for the other side are also indications that they lean towards the opposite direction, although this correlation is less strong. The categories were theoretically determined, but were later empirically confirmed by a factor analysis. The goal is to look into the political cohesion of the UK

(20)

19 parliament with polarization as the measurement of political cohesion. Lower polarization would indicate a higher level of political cohesion and vice versa. If the level of polarization increases then the parties would move away from the center and towards the extremes. If there is no strong movement overall to the extremes or even a movement towards the center then this would indicate that the political cohesion is getting stronger or at least not much weaker between 2015 and 2019.

The MARPOR encourages its users to adapt the RILE to ‘fit their own research purposes’ (Budge I. , The Standard Right-Left Scale, 2013) and in this thesis that results in a scale adjusted around the Brexit situation. However, the adjusted scale is still closely related to the standard RILE scale and will still generally reflect the original right-left dynamic. The specific reasons for which categories are kept, removed and added, and why they are placed right or left are explained in detail in chapter 3.3.4.

The RILE-index consists of 57 policy categories, formulated by the predecessors of the

‘Manifesto Research on Political Representation’ called the ‘Manifesto Research Group’ and the

‘Comparative Manifesto Project’. These categories are used to classify the (quasi-)sentences in the manifesto’s or other selected texts. Some of these categories have sub-categories for more clarity about what exactly falls under the category or to allow easier adjustments to suit ones need when using the data. The classifying is done based references to a topic and not the pro or con position. If a (quasi-)sentence refers to one of the categories then it is marked as such in the table. The categories are based on issues in British and American party platforms between 1920 and 1976, further refined in Manifesto Research Group discussions around 1980 for the purpose of applying them to 20 post- war democracies in Western Europe (Budge & Meyer, 2013) (Robertson, 1976) (Budge & Farlie, 1977, pp. 421-433). In the traditional RILE scale 26 categories are divided between ‘right’ and ‘left’, Each side having thirteen categories. The issues are mostly economic, but also include social, democratic, law & order, and traditional morals issues. That these issues are still valid is shown through the ability of the scale to produce the accurate position of parties across time and states (Budge, et al., 2001) (Klingemann, Volkens, Bara, Budge, & McDonald, 2006). The 26 categories included in the standard RILE-Index are shown in table 1 below.

Table 1: The standard RILE-index

Left emphases sum of % for Right emphases: sum of % for 103 Anti-Imperialism 104 Military: positive

105 Military: Negative 201 Freedom and Human Rights

106 Peace 203 Constitutionalism: Positive

107 Internationalism: Positive 305 Political Authority

202 Democracy 401 Free Market Economy

403 Market Regulation 402 Economic Incentives 404 Economic Planning 407 Protectionism: Negative 406 Protectionism: Positive 414 Economic Orthodoxy 412 Controlled Economy 505 Welfare State Limitation 413 Nationalisation 601 National Way of Life: Positive 504 Welfare State Expension 603 Traditional Morality: Positive 506 Education Expansion 605 Law and Order

701 Labour Groups: Positive 606 Civic Mindedness: Positive

The categories not being classified under right or left will still play a role in the calculation. After processing the manifesto every (quasi-)sentence has a (sub)-category attached to it, if no appropriate

The role of internal cohesion in the UK Parliament and its possible effects on the Brexit negotiations